mlsanityData to 14 Oct 2026

ChronoLens: Measuring Language Change Across Time, Languages, and Linguistic Levels

arXiv 2608.03507 ↗First posted 4 Aug 2026cs.CL cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Historical language change affects morphology, syntax, semantics, and pragmatics, yet computational studies typically examine these levels with incompatible representations and therefore cannot determine whether they evolve together across languages. We address this problem by asking how the magnitude and direction of change vary across linguistic levels, languages, and historical periods within a single analytical space. We introduce ChronoLens, a framework that combines frozen multilingual language models, feature-aligned crosscoders, and post-hoc linguistic interventions, and apply it to 44.98 million documents and approximately 17.2 billion tokens from five parliamentary traditions spanning 1803--2026. The resulting sparse representations agree substantially more strongly with linguistic statistics than dense embeddings or a pooled sparse autoencoder (ρ=0.72 versus 0.29 and 0.28), and reveal that morphology, syntax, semantics, and pragmatics generally change by comparable amounts within a language, while languages differ markedly in when, how far, and in which direction they change. These findings show that historical language change is a structured, multidimensional process: similar magnitudes can conceal different trajectories, and meaningful cross-linguistic comparison requires measuring both distance and direction.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 31 of its 70 references that are in the map. First posted on arXiv 4 Aug 2026.

Builds on

31 of 70 references.

References · 31

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,420
  3. 2020Stanza: A Python Natural Language Processing Toolkit for Many Human Languagescited by 723
  4. 2019Designing and Interpreting Probes with Control Taskscited by 640
  5. 2020Universal Dependencies v2: An Evergrowing Multilingual Treebank Collectioncited by 271
  6. 2016Diachronic Word Embeddings Reveal Statistical Laws of Semantic Changecited by 221
  7. 2026Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnetcited by 191
  8. 2025SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretabilitycited by 124
  9. 2025Are Sparse Autoencoders Useful? A Case Study in Sparse Probingcited by 115
  10. 2024EuroLLM: Multilingual Language Models for Europecited by 106
  11. 2020Analysing Lexical Semantic Change with Contextualised Word Representationscited by 92
  12. 2025Sparse Autoencoders Do Not Find Canonical Units of Analysiscited by 88
19 more
  1. 2019Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsingcited by 70
  2. 2016Cultural Shift or Linguistic Drift? Comparing Two Computational Measures of Semantic Changecited by 63
  3. 2024Why do language models perform worse for morphologically complex languages?cited by 62
  4. 2024A Systematic Comparison of Contextualized Word Embeddings for Lexical Semantic Changecited by 21
  5. 2021Diachronic Analysis of German Parliamentary Proceedings: Ideological Shifts through the Lens of Political Biasescited by 13
  6. 2026Cross-Architecture Model Diffing with Crosscoders: Unsupervised Discovery of Differences Between LLMscited by 12
  7. 2017On the Linearity of Semantic Change: Investigating Meaning Variation via Dynamic Graph Modelscited by 11
  8. 2022Crosslinguistic word order variation reflects evolutionary pressures of dependency and information localitycited by 10
  9. 2017Joint Semantic Synthesis and Morphological Analysis of the Derived Wordcited by 6
  10. 2022Fine-Grained Detection of Solidarity for Women and Migrants in 155 Years of German Parliamentary Debatescited by 6
  11. 2025Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretrainingcited by 6
  12. 2025Narrative Media Framing in Political Discoursecited by 5
  13. 2025Train One Sparse Autoencoder Across Multiple Sparsity Budgets to Preserve Interpretability and Accuracycited by 4
  14. 2025Sparse Autoencoders Can Capture Language-Specific Concepts Across Diverse Languagescited by 4
  15. 2019Target Based Speech Act Classification in Political Campaign Textcited by 3
  16. 2024Quantifying Lexical Semantic Shift via Unbalanced Optimal Transportcited by 3
  17. 2025Framing Migration: A Computational Analysis of UK Parliamentary Discoursecited by 1
  18. 2026DHPLT: large-scale multilingual diachronic corpora and word representations for semantic change modellingcited by 1
  19. 2026HistLens: Mapping Idea Change across Concepts and Corporacited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Chronologic: Measuring Language Models' Ability to Represent the Past
  2. 2026Time Travel Engine: A Shared Latent Chronological Manifold Enables Historical Navigation in Large Language Models
  3. 2025Losing our Tail, Again: (Un)Natural Selection & Multilingual LLMs
  4. 2025IMPACT: Inflectional Morphology Probes Across Complex Typologies
  5. 2026Rethinking Metrics for Lexical Semantic Change Detection
  6. 2026Multilingual Language Models Encode Script Over Linguistic Structure
  7. 2024Remember This Event That Year? Assessing Temporal Information and Reasoning in Large Language Models
  8. 2025How Language Directions Align with Token Geometry in Multilingual LLMs